Topics in Catalysis

, Volume 61, Issue 1–2, pp 119–125 | Cite as

Characterization of Cerium and Oxygen Atoms in Free Clusters of Cerium Oxide by X-ray Absorption Spectroscopy

  • Tetsuichiro Hayakawa
  • Masashi Arakawa
  • Shun Sarugaku
  • Kota Ando
  • Kenichirou Tobita
  • Yuya Kiyomura
  • Tomoki Kawano
  • Akira Terasaki
Original Paper


X-ray absorption spectroscopy of size-selected cerium oxide cluster ions, Ce3O4–7 +, has been carried out by fragment-ion-yield measurement. X-ray absorption spectra measured in the Ce M4-edge and the O K-edge regions provided novel experimental data for chemical analysis of the constituent atoms and for discussion of geometric structures. The spectra near the Ce M4-edge exhibited a clear main peak and subtle substructures. Composition dependence of the spectra indicated that the oxidation state of Ce atoms evolves from + 3 to + 4 as an O atom is introduced one by one. On the other hand, the O K-edge spectra consisted of a main peak at ca. 532 eV and two additional peaks in the pre-edge region, where the intensities of the pre-edge peaks were significantly dependent on Ce:O composition. These pre-edge peaks suggested that the geometric structures of Ce3O5–7 + retain the “framework” of Ce3O4 + with “peripheral” O atoms bound on-top to the Ce atoms; coexistence of a structural isomer was suggested for Ce3O6 +, indicating presence of a molecular oxygen with Ce3O4 +−O2 geometry.


Cerium oxide clusters X-ray absorption spectroscopy Chemical analysis Oxidation states Composition dependence 

1 Introduction

Among catalysts for redox processes, cerium oxide has been attracting much attention as an important material in industry, which is employed in many applications such as automobile exhaust catalysts [1]. The remarkable property of cerium oxides is oxygen storage/release due to flexibility of the oxidation state of cerium between Ce(III) and Ce(IV) [2]. The oxidation state of cerium is considered to be relevant to its catalytic activity such as NO reduction [3, 4]. In this context, investigation of the oxidation state of cerium is important for characterizing cerium-oxide catalysts.

To gain insights into mechanisms of catalysis, studies of isolated clusters have an advantage in investigating local chemical properties in the vicinity of active sites; clusters provide us with a model system of catalysts. From this point of view, reaction of isolated cerium-oxide clusters has been investigated to understand the oxygen storage/release behavior and the catalytic activity of cerium oxides. Wu et al. studied reaction of positively-charged Ce m O n + (m = 2–6, n = 2 m) with hydrocarbons and CO [5] and that of negatively-charged Ce m O n (m = 4–21, n = 2m + 1) with CO [6]. Hirabayashi and Ichihashi studied reactivity of Ce m O n + (m = 2–6) with CO, NO, and O2 molecules as a function of oxygen content, n [7, 8]. Nagata et al. reported temperature-dependence as well as composition-dependence of the reaction of Ce m O n + (m = 2–9) with CO, NO, and several other gaseous molecules [9, 10, 11]. However, reactivity of cerium-oxide clusters has been discussed only on the basis of a formal average oxidation number of the Ce atoms as well as geometric structures, total energies, and charges on each Ce atom obtained computationally. This is due to lack of fundamental experimental data related to geometric and electronic structures. As for such experimental studies, Burow et al. reported geometric structures of Ce m O n + by vibrational spectroscopy [12]. They claimed that even small clusters (m = 3–5) retain structural motifs typical of bulk CeO2. Akin et al. studied stability and geometry of Ce m O n + (m = 3–11) by photodissociation [13].

Experimental studies on electronic properties are also required. Direct measurement of the charge state of constituent atoms in the clusters, for example, allows us to discuss relationship between reactivity and the charge state. For bulk cerium oxides, one finds many reports on the charge state of Ce atoms measured by X-ray absorption spectroscopy (XAS) [14, 15] and electron-energy loss spectroscopy (EELS) [16, 17], i.e., inner-shell excitation spectroscopy. The element-specific feature of these spectroscopic techniques provides us with powerful means to investigate local information on geometries and electronic states in multi-element materials.

Application of XAS to isolated clusters, however, has a difficulty due to the low density of samples as well as small absorption cross sections of inner-shell excitation compared with those of valence excitation. Therefore, XAS of free size-selected clusters has been developed step by step. First, XAS of isolated clusters has been demonstrated without size-selection in 1990s [18, 19, 20]. Second, an attempt to size-selection was made for clusters of chalcogens, which were able to be generated with specific size distribution (quasi-size-selection) [21, 22]. And more recently, an ion-trap technique enabled XAS of size-selected cluster ions [23, 24].

In this context, we have developed an experimental apparatus for XAS equipped with a linear RF ion trap, which is transportable to synchrotron X-ray facilities. The first experiment was performed on size-selected cerium oxide cluster ions, Ce2O3 + and Ce2O5 +, where XAS spectra were successfully measured near the Ce M-edge, enabling discussion of oxidation states of Ce atoms [25]. The present study extends our first study to a series of Ce3O n + (n = 4–7) cluster ions to characterize both Ce and O atoms by measuring the O K-edge as well as the Ce M-edge region. Measurement at a new X-ray beamline provides an energy resolution higher than our previous measurement, which shows an advantage in observing detailed structures in the spectra. The systematic change in the oxygen content from n = 4–7 reveals evolution of the oxidation state of Ce atoms and the chemical environment of O atoms as the oxygen content is altered.

2 Experimental Procedures

Experimental details have been described elsewhere [25]. Our apparatus consists of a magnetron-sputtering cluster ion source, octopole ion guides, a quadrupole mass filter for size selection, a linear quadrupole RF ion trap, and a time-of-flight (TOF) mass spectrometer for fragment-ion detection. A quadrupole type was adopted for the ion trap, rather than octopole, to obtain better overlap between X-ray beam and stored ions [26, 27, 28]. Cerium-oxide cluster ions were produced by introducing a mixed gas of helium and oxygen into an aggregation chamber of the cluster ion source cooled by liquid nitrogen. The cluster ions were transported by subsequent ion guides and a quadrupole deflector and were admitted into a quadrupole mass filter for size-selection. Cluster ions thus size-selected were deflected by a second quadrupole deflector and were introduced into an ion trap located on the axis of the X-ray beam. The linear ion trap was surrounded by a jacket cooled by liquid nitrogen and was filled with a buffer He gas at about 0.1 Pa. The ions thus stored were irradiated with soft X-ray from a synchrotron tunable at the Ce M-edge around 900 eV and the O K-edge around 530 eV. As cluster ions undergo fragmentation upon X-ray absorption, an X-ray absorption spectrum was recorded through the fragment-ion yield as a function of photon energy (the fragment-ion yield mode). The fragment ions produced in the present experiment were CeO2 +, CeO+, Ce+ and Ce2+, of which the yields were added to derive a total fragment-ion yield.

XAS was carried out at the beamline BL-2B of the synchrotron facility Photon Factory (KEK, Tsukuba, Japan) after preliminary measurement at BL-7A. The synchrotron radiation in BL-2B was from an undulator designed for the soft X-ray region (250–2000 eV), which was monochromatized by a variable-included-angle Monk-Gillieson monochromator with a varied-line-spacing plane grating [29]. The X-ray beam size at the ion trap was ca. 0.5 and 2 mm for vertical and horizontal, respectively. The photon-energy resolution was about 0.5 eV in the vicinity of 900 eV, when the photon flux was kept at about 1 × 1012 photons/s.

XAS spectra of CeO2 and CeF3 powders were measured as well in the Ce M-edge region as references for Ce(IV) and Ce(III), respectively; calibration of the photon energy was carried out at the same time. XAS of a Cr2O3 powder in the Cr L-edge region was also measured for calibration of the photon energy near the O K-edge. XAS spectra of these powder samples were measured in a total-electron-yield mode by recording a drain current from the sample fixed on a carbon adhesive tape.

3 Results

3.1 Mass Spectrum of Cerium Oxide Clusters

A typical mass spectrum of cerium oxide cluster ions, Ce m O n +, is shown in Fig. 1. Ce3O n + (n = 4–7) were produced dominantly in the series of m = 3 at a sufficient amount of oxygen flow, whereas Ce3O3 + and Ce3O8 + were hardly observed. Ce4O6–10 + and Ce5O8–12 + were the major species observed for m = 4 and 5. This size-distribution was similar to those of previous reports [5, 13].

Fig. 1

A mass spectrum of cerium oxide cluster ions, Ce m O n +, optimized for producing m = 3 with n = 4–7

3.2 Reference Spectra

Figure 2 shows XAS spectra of CeO2 and CeF3 powders in the Ce M-edge region measured for references of Ce(IV) and Ce(III), respectively. Two main peaks corresponding to M5 and M4 absorption are clearly seen along with additional substructures. The main peak of the CeF3 spectrum is observed at 899.9 eV in the M4 region with substructures at 896.8 and 898.5 eV. In the M4 peak of the CeO2 spectrum, on the other hand, the main peak is located at 901.8 eV with a satellite at 907.1 eV; two subtle peaks and a shoulder discernible at 896.8, 898.5, and 899.9 eV, respectively, indicate that a small amount of Ce(III) exists in the present CeO2 sample probably due to partial release of oxygen. These spectra are in good agreement with previous reports [30, 31].

Fig. 2

X-ray absorption spectra of reference samples, i.e., CeF3 and CeO2 powders

3.3 X-ray Absorption Spectra near Cerium M-edge

Figure 3 shows XAS spectra of Ce3O4–7 + measured in the fragment-ion-yield mode in the Ce M4-edge region. The spectra are normalized by the edge jump from 890 to 910 eV. The spectrum of Ce3O4 + shows a clear main peak at the energy of 899.8 eV and substructures at 896.8 and 898.3 eV. Those of Ce3O6 + and Ce3O7 + are almost identical with each other, which exhibit a main peak at 901.1 eV and a satellite peak at 906.8 eV. The spectrum of Ce3O5 + is characterized by the features of both Ce3O4 + and Ce3O6, 7 +, i.e., substructures at 896.8 and 898.3 eV and a satellite peak at 906.8 eV.

Fig. 3

X-ray absorption spectra of Ce3O4–7 + near the Ce M4-edge. Results of data fitting are superimposed by solid curves

The curves shown in Fig. 3 were obtained by fitting the spectra with Gaussian profiles for the observed peaks and an error function for the edge jump. For Ce3O6 + and Ce3O7 +, the spectra were reproduced by an identical curve. The spectrum of Ce3O5 + was able to be reproduced by a linear combination of the two fitting curves for Ce3O4 + and Ce3O6 +, i.e.,
$$I\! \left( {{\rm Ce_3}{\rm O_5}^{+}} \right)=\frac{1}{3}\ I\! \left( {{\rm Ce_3}{\rm O_4}^{+}} \right)+\frac{2}{3}\ I\! \left( {{\rm Ce_3}{\rm O_6}^{+}} \right)$$

3.4 X-ray Absorption Spectra near Oxygen K-edge

Figure 4 shows XAS spectra of Ce3O4–7 + in the O K-edge region. There are three features in the spectra: a main peak at 532.0–533.5 eV and additional pre-edge peaks A and B at 529.5 and 526.8 eV, respectively. The main peak is located at 532.0 eV for Ce3O4 +, which shifts toward higher energies with the addition of oxygen and reaches 533.5 eV for Ce3O7 +. The pre-edge peak A becomes more intense without energy shifts as the oxygen content increases from n = 5 to 7, whereas it is not present in the spectrum of Ce3O4 +. The peak B shows up only for Ce3O6 +.

Fig. 4

X-ray absorption spectra of Ce3O4–7 + near the O K-edge. Results of data fitting are superimposed by solid curves

4 Discussion

4.1 Stability of Cerium Oxide Clusters

The size distribution of the clusters shown in Fig. 1 clearly indicates that Ce m O n + (n = 4–7) are stable, whereas Ce3O3 + and Ce3O8 + are significantly unstable. The overall feature in the relative stability is in agreement with previous studies [5, 13]. The variation in the size distribution may be caused by experimental procedures or by an amount of oxygen introduced upon cluster formation. Several computational studies have reported geometric structures of Ce3O n + [5, 6, 8, 10, 12, 13, 32]. Geometric structures of Ce3O n + predicted by the calculations are illustrated in Fig. 5. Calculations predicted that the most stable structure of Ce3O4 + is 3-4 [10, 12], and that Ce3O5–7 + are formed by additional O atoms bound to the Ce3O4 + “framework”. For Ce3O5 +, two different structures 3-5a [8, 12, 32] and 3-5b [10, 13] are suggested by computational studies, among which 3-5a is supported experimentally by vibrational spectroscopy [12]. For Ce3O6 +, 3-6a is suggested by most of the calculations [5, 8, 32]; 3-6b is the second stable isomer, where a molecular oxygen is bound to Ce3O4 + according to the unpublished computational result by Ichihashi. 3-7 is the most stable structure calculated for Ce3O7 [6]. The high stability of Ce3O4–7 + in contrast to Ce3O3 + and Ce3O8 + suggests that Ce3O5–7 + are formed on the basis of the Ce3O4 + “framework” with “peripheral” O atoms attached one by one. From a catalytic point of view, Ce3O4–7 + can be considered as a model system of oxygen storage/release materials, where Ce3O4 + is the fully reduced state and Ce3O7 + is the fully oxidized state.

Fig. 5

Geometric structures of Ce3O n + predicted by calculations. Ce: yellow, O: red. 3-4 [10, 12] is optimized geometry for Ce3O4 +, 3-5a [8, 12, 32] and 3-5b [10, 13] are for Ce3O5 +, 3-6a [5, 8, 32] is optimized geometry and 3-6b is the low energy isomer of Ce3O6 + calculated by M. Ichihashi, 3-7 [6] is optimized geometry of Ce3O7

4.2 Charge State of Cerium Atoms in the Clusters

The XAS spectrum of Ce3O4 + in the Ce M4 region is almost identical with that of the CeF3 powder (see Figs. 2, 3). This agreement indicates that all the Ce atoms are in the oxidation state of + 3, which corresponds to the formal average oxidation number of Ce calculated for Ce3O4 +. For Ce3O6 + and Ce3O7 +, on the other hand, the XAS spectra in Fig. 3 show that the oxidation number of all the Ce atoms is + 4, which is smaller than the formal average oxidation numbers + 4.3 and + 5.0 for Ce3O6 + and Ce3O7 +, respectively. This result implies that the oxidation state of the Ce atoms reaches a plateau at + 4, as in the bulk cerium oxides. Finally, the spectrum of Ce3O5 + is reproduced well by the curve expressed by Eq. (1). This means that one Ce atom is Ce(III) and the other two are Ce(IV), as is consistent with the formal average oxidation number + 3.7 for Ce3O5 +. It is thus concluded that the oxidation state of the Ce atoms in Ce3O n + changes from + 3 to + 4 upon increase in the number of oxygen atoms from n = 4 to 6.

4.3 Chemical State of Oxygen Atoms

All the spectra in Fig. 4 measured in the O K-edge commonly show a main peak in the energy range of 532–533 eV. It suggests that the main peak originates from the O atoms in the Ce3O4 + “framework”; the peak is probably caused by excitation of 1 s electrons of the O atoms into an unoccupied state between the O and Ce atoms. Bonding character of the unoccupied state can be speculated from the fragment-ion distribution as discussed in the previous report [25]. In the present case, the increased branching fraction of the molecular CeO+ fragment at the main peak region suggests that the main peak corresponds to excitation to anti-bonding molecular orbital, as discussed in the Supplementary Material. Since a peak shift is generally caused by a change in the charge state of the atom, the blue shift observed for the main peak from 532 eV (Ce3O4 +) to 533 eV (Ce3O7 +) implies that the local charge on the O atoms in the “framework” becomes less negative when more and more “peripheral” O atoms are bound. This interpretation of the peak shift is consistent with the structural model illustrated in Fig. 5.

The pre-edge peaks, A and B, exhibit significant dependence on the number of O atoms, n. Absence of these peaks in the spectrum of Ce3O4 + implies that they do not originate from the “framework” O atoms. As for peak B, a similar peak (at ca. 8 eV lower than the main peak) has been reported in the O K-edge EELS spectra of several solid metal oxides [33, 34]. They assigned it to the π* peak of the O–O bond of oxygen molecules produced by defect formation upon electron irradiation. This assignment rationalizes our interpretation that isomer 3-6b coexists in the present Ce3O6 +, which has a molecular oxygen attached in the form of Ce3O4 +−O2, whereas the other compositions with n = 4, 5, and 7 do not possess O2 and, therefore, peak B is absent. Oxygen storage by adsorption of O2 molecules was predicted by calculation [35]. Existence of similar molecular oxygen was also reported for larger nanoparticles, which may lead to dramatic enhancement of oxygen-storage capacity of cerium oxide [36, 37].

Peak A, on the other hand, is assigned to excitation of 1 s electrons of O atoms into unoccupied orbital originating from O 2p hybridized with Ce 4f or 5d. A peak at a similar energy observed for bulk cerium oxide is assigned to excitation into O 2p hybridized with Ce 4f orbital [38, 39]. Observation of peak A in Ce3O5–7 + is explained as follows: As the energy of the hybridized orbital is dependent on the bond length between O and Ce atoms, the appearance of peak A at an energy lower than the main peak implies that the bond length of the relevant O atoms is shorter than the Ce–O bond in the Ce3O4 + “framework”. In this context, the bond length between the “peripheral” O atom and the nearest Ce atom has been shown to be shorter than those in the “framework” by a computational study [8]. Therefore, it is likely that peak A originates from the “peripheral” O atoms in Ce3O5–7 + on the premise that the bonding orbital would shift to sufficiently lower energy than the main peak.

As the pre-edge peaks A and B are ascribable to such “peripheral” or molecular oxygen, the total intensity of the pre-edge peaks (I A + I B) should be related to geometric structures of the clusters when it is compared with that of the main peak (I main) assigned to “framework” oxygen. Figure 6 displays the relative intensity of the pre-edge peaks, (I A + I B)/I main, obtained by fitting the O K-edge spectra, as a function of the number n of oxygen atoms in the cluster. It shows monotonic increase with n, indicating that the number of “peripheral” O atoms increases as an O atom is introduced one by one to Ce3O4 + up to Ce3O7 +. This result is consistent with calculated geometries of the clusters as follows: the number of “peripheral” oxygen atoms is zero, one, and three in Ce3O4 + (3–4), Ce3O5 + (3-5a), and Ce3O7 + (3-7), respectively, while the number of “framework” O atoms stays constant. For Ce3O6 +, peak A is not as intense as inferred from two “peripheral” O atoms in the cluster (3-6a). This is explained by coexistence of isomer 3-6b; the combined intensity of peaks A and B follows the general trend as shown in Fig. 6.

Fig. 6

Relative intensity (I A + I B)/I main, where I A, I B and I main are the intensities of peaks A, B, and the main peak, respectively

From the point of view of oxygen storage/release, the present study reveals that oxygen can be stored as atoms or dimers in the periphery of the clusters. This oxygen storage mechanism of the cluster is different from that of bulk cerium oxide, where the surface contribution is much smaller than clusters. In this context, we found that the clusters can store oxygen even when the formal oxidation number of Ce atoms exceeds + 4, i.e., Ce3O6 + and Ce3O7 +. This finding suggests that cerium oxide clusters would be able to store more oxygen atoms per Ce atom than bulk cerium oxide.

5 Conclusions

A series of size-selected cerium-oxide cluster ions, Ce3O n +, were studied systematically in the oxygen-content range of n = 4–7 as a model system of oxygen storage/release materials. Chemical analysis of the constituent Ce and O atoms was carried out by XAS measured both in the Ce M-edge and the O K-edge regions. Due to the improved resolution in the new beamline in KEK-PF, which allowed us to observe even subtle structures, the Ce M4-edge spectra provided clear evidence of the oxidation state of the Ce atoms in each cluster. The oxidation number evolved from + 3 to + 4 as the oxygen content increased from n = 4 to 6 without a further change for n = 7. The O K-edge spectra, on the other hand, exhibited a main peak and additional pre-edge peaks. These peaks provided a clue to characterize O atoms into three types: the first is those in the Ce3O4 + “framework” contributing to the main peak, the second is the “peripheral” O atoms bound on-top to the Ce atoms in the “framework” giving rise to peak A in the pre-edge, and the third is a molecular oxygen present in one of the isomers of Ce3O6 +. The present XAS spectra thus provide not only a plausible scenario of oxidation of Ce3O4 + to Ce3O7 + but also novel experimental evidence for evaluating the structural models proposed by computational studies, demonstrating a powerful technique characterizing catalytic materials in the cluster regime.



We thank Professors Hiroshi Kumigashira, Koji Horiba, and Makoto Minohara for their technical support in the beamline BL-2B, KEK-PF, and Professor Kenta Amemiya in the beamline BL-7A. We are grateful to Professor Masahiko Ichihashi for informing us about his computational results prior to publication. We also thank Dr. Kazuhiro Egashira for his help in preparatory experiments. This work has been supported by the Special Cluster Research Project of Genesis Research Institute, Inc., and was performed under the approval of the Photon Factory Program Advisory Committee (Proposal No. 2014G091 and 2016G183).

Supplementary material

11244_2017_869_MOESM1_ESM.docx (169 kb)
Supplementary material 1 (DOCX 168 KB)


  1. 1.
    Trovarelli A, Fornasiero P (2013) Catalysis by Ceria and related materials. 2nd edn. Catalytic science series, vol 12. Imperial College Press, LondonGoogle Scholar
  2. 2.
    Skorodumova NV, Simak SI, Lundqvist BI, Abrikosov IA, Johansson B (2002) Quantum origin of the oxygen storage capability of ceria. Phys Rev Lett 89:166601CrossRefGoogle Scholar
  3. 3.
    Liu L, Cao Y, Sun W, Yao Z, Liu B, Gao F, Dong L (2011) Morphology and nanosize effects of ceria from different precursors on the activity for NO reduction. Catal Today 175:48–54CrossRefGoogle Scholar
  4. 4.
    Nolan M, Parker SC, Watson GW (2006) CeO2 catalyzed conversion of CO, NO2 and NO from first principles energetics. Phys Chem Chem Phys 8:216–218CrossRefGoogle Scholar
  5. 5.
    Wu X, Zhao Y, Xue W, Wang Z, He S, Ding X (2010) Active sites of stoichiometric cerium oxide cations (CemO2 m +) probed by reactions with carbon monoxide and small hydrocarbon molecules. Phys Chem Chem Phys 12:3984–3997CrossRefGoogle Scholar
  6. 6.
    Wu X, Ding X, Bai S, Xu B, He S, Shi Q (2011) Experimental and theoretical study of the reactions between cerium oxide cluster anions and carbon monoxide: size-dependent reactivity of CenO2n+1 (n=1 − 21). J Phys Chem C 115:13329–13337CrossRefGoogle Scholar
  7. 7.
    Hirabayashi S, Ichihashi M (2013) Oxidation of composition-selected cerium oxide cluster cations by O2. Chem Phys Lett 564:16–20CrossRefGoogle Scholar
  8. 8.
    Hirabayashi S, Ichihashi M (2013) Oxidation of CO and NO on composition-selected cerium oxide cluster cations. J Phys Chem A 117:9005–9010CrossRefGoogle Scholar
  9. 9.
    Nagata T, Miyajima K, Mafune F (2015) Stable stoichiometry of gas-phase cerium oxide cluster ions and their reactions with CO. J Phys Chem A 119:1813–1819CrossRefGoogle Scholar
  10. 10.
    Nagata T, Miyajima K, Hardy RA, Metha GF, Mafune F (2015) Reactivity of oxygen deficient cerium oxide clusters with small gaseous molecules. J Phys Chem A 119:5545–5552CrossRefGoogle Scholar
  11. 11.
    Nagata T, Miyajima K, Mafune F (2015) Oxidation of nitric oxide on gas-phase cerium oxide clusters via reactant adsorption and product desorption processes. J Phys Chem A 119:10255–10263CrossRefGoogle Scholar
  12. 12.
    Burow AM, Wende T, Sierka M, Włodarczyk R, Sauer J, Claes P, Jiang L, Meijer G, Lievens P, Asmis KR (2011) Structures and vibrational spectroscopy of partially reduced cerium oxide clusters. Phys Chem Chem Phys 13:19393–19400CrossRefGoogle Scholar
  13. 13.
    Akin ST, Ard SG, Dye BE, Schaefer HF, Duncan MA (2016) Photodissociation of cerium oxide nanocluster cations. J Phys Chem A 120:2313–2319CrossRefGoogle Scholar
  14. 14.
    Niewa R, Hu Z, Grazioli C, Rossler U, Golden MS, Knupfer M, Fink J, Giefers H, Wortmann G, de Groot FMF, DiSalvo FJ (2002) XAS spectra of Ce2[MnN3] at the Ce-M4,5, Ce-L3, Mn-L2,3 and N-K thresholds. J Alloys Comp 346:129–133CrossRefGoogle Scholar
  15. 15.
    Chen S, Lu Y, Huang T, Yan D, Dong C (2010) Oxygen vacancy dependent magnetism of CeO2 nanoparticles prepared by thermal decomposition method. J Phys Chem 114:19576–19581Google Scholar
  16. 16.
    Garvie LAJ, Buseck PR (1999) Determination of Ce4+/Ce3+ in electron-beam-damaged CeO2 by electron energy-loss spectroscopy. J Phys Chem Solids 60:1943–1947CrossRefGoogle Scholar
  17. 17.
    Wu L, Wiesmann J, Moodenbaugh AR, Klie RF, Zhu Y, Welch DO, Suenaga M (2004) Oxidation state and lattice expansion of CeO2–x nanoparticles as a function of particle size. Phys Rev B 69:125415CrossRefGoogle Scholar
  18. 18.
    Rühl E, Jochims HW, Schmale C, Biller E, Hitchcock AP, Baumgärtel H (1991) Core-level excitation in argon clusters. Chem Phys Lett 178:558–564CrossRefGoogle Scholar
  19. 19.
    Rühl E, Heinzel C, Hitchcock AP, Baumgärtel H (1993) Ar 2p spectroscopy of free argon clusters. J Chem Phys 98:2653–2663CrossRefGoogle Scholar
  20. 20.
    Björneholm O, Federmann F, Joppien M, Fössing F, Kakar S, von Pietrowski R, Möller T (1996) Valence-and inner-shell spectroscopy on rare-gas clusters. Surf Rev Lett 3:299–306CrossRefGoogle Scholar
  21. 21.
    Hayakawa T, Nagaya K, Hamada K, Ohmasa Y, Yao M (2000) Photoelectron photoion coincidence measurements of selenium cluster beam. II. photon energy dependence. J Phys Soc Jpn 69:2850–2858CrossRefGoogle Scholar
  22. 22.
    Nagaya K, Yao M, Hayakawa T, Ohmasa Y, Kajihara Y, Ishii M, Katayama Y (2002) Size-selective extended X-ray absorption fine structure spectroscopy of free selenium clusters. Phys Rev Lett 89:243401CrossRefGoogle Scholar
  23. 23.
    Lau JT, Rittmann J, Zamudio-Bayer V, Vogel M, Hirsch K, Klar Ph, Lofink F, Möller T (2008) Size dependence of L2,3 branching ratio and 2p core-hole screening in X-ray absorption of metal clusters. Phys Rev Lett 101:153401CrossRefGoogle Scholar
  24. 24.
    Hirsch K, Lau JT, Klar P, Langenberg A, Probst J, Rittmann J, Vogel M, Zamudio-Bayer V, Möller T, von Issendorff B (2009) X-ray spectroscopy on size-selected clusters in an ion trap: from the molecular limit to bulk properties. J Phys B 42:154029CrossRefGoogle Scholar
  25. 25.
    Hayakawa T, Egashira K, Arakawa M, Ito T, Sarugaku S, Ando K, Terasaki A (2016) X-ray absorption spectroscopy of Ce2O3 + and Ce2O5 + near Ce M-edge. J Phys B 49:075101CrossRefGoogle Scholar
  26. 26.
    Terasaki A, Majima T, Kondow T (2007) Photon-trap spectroscopy of mass-selected ions in an ion trap: optical absorption and magneto-optical effects. J Chem Phys 127:231101CrossRefGoogle Scholar
  27. 27.
    Terasaki A, Majima T, Kasai C, Kondow T (2009) Photon-Trap Spectroscopy of size-selected free cluster ions: “direct” measurement of optical absorption of Ag9 +. Eur Phys J D 52:43–46CrossRefGoogle Scholar
  28. 28.
    Majima T, Santambrogio G, Bartels C, Terasaki A, Kondow T, Meinen J, Leisner T (2012) Spatial distribution of ions in a linear octopole radio-frequency ion trap in the space-charge limit. Phys Rev A 85:053414CrossRefGoogle Scholar
  29. 29.
    Construction of new wide-energy range VUV & SX beamline BL-2 “MUSASHI”. Photon Factory Activity Report 2013 #31, Newly Developed Experimental Facilities, pp 1–3Google Scholar
  30. 30.
    Kalkowski G, Kaindl G, Wortmann G, Lentz D, Krause S (1988) 4f-ligand hybridization in CeF4 and TbF4 probed by core-level spectroscopies. Phys Rev B 37:1376–1382CrossRefGoogle Scholar
  31. 31.
    Howald L, Stilp E, de Réotier PD, Yaouanc A, Raymond S, Piamonteze C, Lapertot G, Baines C, Keller H (2015) Evidence for coexistence of bulk superconductivity and itinerant antiferromagnetism in the heavy fermion system CeCo(In1−xCdx)5. Sci Rep 5:12528CrossRefGoogle Scholar
  32. 32.
    Zhou ZX, Wang LN, Li ZY, He SG, Ma TM (2016) Oxidation of SO2 to SO3 by cerium oxide cluster cations Ce2O4 + and Ce3O6 +. J Phys Chem A 120:3843–3848CrossRefGoogle Scholar
  33. 33.
    Jiang N, Spence CH (2006) Interpretation of oxygen K pre-edge peak in complex oxides. Ultramicroscopy 106:215–219CrossRefGoogle Scholar
  34. 34.
    Braaten NA, Borg A, Grepstad JK, Raaen S, Ruckman MW (1991) Oxygen K near-edge-structure for thin Ce oxide films. Solid State Commun 77:731–734CrossRefGoogle Scholar
  35. 35.
    Preda G, Migani A, Neyman KM, Bromley ST, Illas F, Pacchioni G (2011) Formation of superoxide anions on ceria nanoparticles by interaction of molecular oxygen with Ce3+ sites. J Phys Chem C 115:5817–5822CrossRefGoogle Scholar
  36. 36.
    Kullgren J, Hermansson K, Broqvist P (2013) Supercharged low-temperature oxygen storage capacity of ceria at the nanoscale. J Phys Chem Lett 4:604–608CrossRefGoogle Scholar
  37. 37.
    Huang X, Beck MJ (2015) “activated oxygen” molecules on ceria nanoparticles. Chem Mater 27:5840–5844CrossRefGoogle Scholar
  38. 38.
    Douillard L, Gautier M, Thromat N, Henriot M, Guittet MJ, Duraud JP, Tourillon G (1994) Local electronic structure of Ce-doped Y2O3: an XPS and XAS study. Phys Rev B 49:16171–16180CrossRefGoogle Scholar
  39. 39.
    Mullins DR, Overbury SH, Huntley DR (1998) Electron spectroscopy of single crystal and polycrystalline cerium oxide surfaces. Surf Sci 409:307–319CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2017

Authors and Affiliations

  1. 1.East Tokyo LaboratoryGenesis Research Institute, Inc.IchikawaJapan
  2. 2.Department of Chemistry, Faculty of ScienceKyushu UniversityNishi-kuJapan

Personalised recommendations